Tweak ligature component matching for ligature formation

If two marks want to ligate and they belong to different components of the
same ligature glyph, and said ligature glyph is to be ignored according to
mark-filtering rules, then allow.

Example Burmese senquence:

  U+1004,U+103A,U+1039,U+101B,U+103D,U+102D

Test font provided by Norbert Lindenberg.

Fixes https://github.com/behdad/harfbuzz/issues/545
This commit is contained in:
Behdad Esfahbod 2017-10-02 20:02:45 +02:00
parent 71c0a1429d
commit 8b2c94c43f
3 changed files with 51 additions and 9 deletions

View File

@ -374,6 +374,13 @@ struct hb_apply_context_t :
inline void reject (void) { num_items++; match_glyph_data--; }
inline matcher_t::may_skip_t
may_skip (const hb_apply_context_t *c,
const hb_glyph_info_t &info) const
{
return matcher.may_skip (c, info);
}
inline bool next (void)
{
assert (num_items > 0);
@ -736,11 +743,17 @@ static inline bool match_input (hb_apply_context_t *c,
* - Ligatures cannot be formed across glyphs attached to different components
* of previous ligatures. Eg. the sequence is LAM,SHADDA,LAM,FATHA,HEH, and
* LAM,LAM,HEH form a ligature, leaving SHADDA,FATHA next to eachother.
* However, it would be wrong to ligate that SHADDA,FATHA sequence.o
* There is an exception to this: If a ligature tries ligating with marks that
* belong to it itself, go ahead, assuming that the font designer knows what
* they are doing (otherwise it can break Indic stuff when a matra wants to
* ligate with a conjunct...)
* However, it would be wrong to ligate that SHADDA,FATHA sequence.
* There are a couple of exceptions to this:
*
* o If a ligature tries ligating with marks that belong to it itself, go ahead,
* assuming that the font designer knows what they are doing (otherwise it can
* break Indic stuff when a matra wants to ligate with a conjunct,
*
* o If two marks want to ligate and they belong to different components of the
* same ligature glyph, and said ligature glyph is to be ignored according to
* mark-filtering rules, then allow.
* https://github.com/behdad/harfbuzz/issues/545
*/
bool is_mark_ligature = _hb_glyph_info_is_mark (&buffer->cur());
@ -761,13 +774,41 @@ static inline bool match_input (hb_apply_context_t *c,
unsigned int this_lig_id = _hb_glyph_info_get_lig_id (&buffer->info[skippy_iter.idx]);
unsigned int this_lig_comp = _hb_glyph_info_get_lig_comp (&buffer->info[skippy_iter.idx]);
if (first_lig_id && first_lig_comp) {
if (first_lig_id && first_lig_comp)
{
/* If first component was attached to a previous ligature component,
* all subsequent components should be attached to the same ligature
* component, otherwise we shouldn't ligate them. */
* component, otherwise we shouldn't ligate them... */
if (first_lig_id != this_lig_id || first_lig_comp != this_lig_comp)
return_trace (false);
} else {
{
if (first_lig_id != this_lig_id && this_lig_id != 0)
return_trace (false);
/* ...unless, we are attached to a base ligature and that base
* ligature is ignorable. */
bool found = false;
const hb_glyph_info_t *out = buffer->out_info;
unsigned int j = buffer->out_len;
while (j && _hb_glyph_info_get_lig_id (&out[j - 1]) == first_lig_id)
{
if (_hb_glyph_info_get_lig_comp (&out[j - 1]) == 0)
{
j--;
found = true;
break;
}
j--;
}
if (!found)
return_trace (false);
if (skippy_iter.may_skip (c, out[j]) != hb_apply_context_t::matcher_t::SKIP_YES)
return_trace (false);
}
}
else
{
/* If first component was NOT attached to a previous ligature component,
* all subsequent components should also NOT be attached to any ligature
* component, unless they are attached to the first component itself! */

View File

@ -33,3 +33,4 @@ fonts/sha1sum/1c2fb74c1b2aa173262734c1f616148f1648cfd6.ttf::U+0995,U+09CD,U+0995
fonts/sha1sum/1c2fb74c1b2aa173262734c1f616148f1648cfd6.ttf::U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0020,U+0995,U+09CD,U+09B0,U+0995,U+09CD,U+09B2:[u0995_u09CD.half_u0995.pres=0+566|u0995_u09CD.half_u0995.pres=3+566|u0995_u09CD.half_u0995.pres=6+566|u0995_u09CD.half_u0995.pres=9+566|u0995_u09CD.half_u0995.pres=12+566|u0995_u09CD.half_u0995.pres=15+566|u0995_u09CD.half_u0995.pres=18+566|u0995_u09CD.half_u0995.pres=21+566|u0995_u09CD.half_u0995.pres=24+566|u0995_u09CD.half_u0995.pres=27+566|u0995_u09CD.half_u0995.pres=30+566|u0995_u09CD.half_u0995.pres=33+566|u0995_u09CD.half_u0995.pres=36+566|u0995_u09CD.half_u0995.pres=39+566|u0995_u09CD.half_u0995.pres=42+566|u0995_u09CD.half_u0995.pres=45+566|u0995_u09CD.half_u0995.pres=48+566|u0995_u09CD.half_u0995.pres=51+566|u0995_u09CD.half_u0995.pres=54+566|u0995_u09CD.half_u0995.pres=57+566|u0995_u09CD.half_u0995.pres=60+566|u0995_u09CD.half_u0995.pres=63+566|u0995_u09CD.half_u0995.pres=66+566|u0995_u09CD.half_u0995.pres=69+566|u0995_u09CD.half_u0995.pres=72+566|u0995_u09CD.half_u0995.pres=75+566|u0995_u09CD.half_u0995.pres=78+566|u0995_u09CD.half_u0995.pres=81+566|u0995_u09CD.half_u0995.pres=84+566|u0995_u09CD.half_u0995.pres=87+566|u0995_u09CD.half_u0995.pres=90+566|u0995_u09CD.half_u0995.pres=93+566|u0995_u09CD.half_u0995.pres=96+566|space=99+213|u0995_u09B0_u09CD.blwf.vatu=100+643|u0995_u09CD.half_u09B2.pres=103+602]
fonts/sha1sum/1c2fb74c1b2aa173262734c1f616148f1648cfd6.ttf::U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0020,U+0995,U+09CD,U+09B0,U+0995,U+09CD,U+09B2:[u0995_u09CD.half_u0995.pres=0+566|u0995_u09CD.half_u0995.pres=3+566|u0995_u09CD.half_u0995.pres=6+566|u0995_u09CD.half_u0995.pres=9+566|u0995_u09CD.half_u0995.pres=12+566|u0995_u09CD.half_u0995.pres=15+566|u0995_u09CD.half_u0995.pres=18+566|u0995_u09CD.half_u0995.pres=21+566|u0995_u09CD.half_u0995.pres=24+566|u0995_u09CD.half_u0995.pres=27+566|u0995_u09CD.half_u0995.pres=30+566|u0995_u09CD.half_u0995.pres=33+566|u0995_u09CD.half_u0995.pres=36+566|u0995_u09CD.half_u0995.pres=39+566|u0995_u09CD.half_u0995.pres=42+566|u0995_u09CD.half_u0995.pres=45+566|u0995_u09CD.half_u0995.pres=48+566|u0995_u09CD.half_u0995.pres=51+566|u0995_u09CD.half_u0995.pres=54+566|u0995_u09CD.half_u0995.pres=57+566|u0995_u09CD.half_u0995.pres=60+566|u0995_u09CD.half_u0995.pres=63+566|u0995_u09CD.half_u0995.pres=66+566|u0995_u09CD.half_u0995.pres=69+566|u0995_u09CD.half_u0995.pres=72+566|u0995_u09CD.half_u0995.pres=75+566|u0995_u09CD.half_u0995.pres=78+566|u0995_u09CD.half_u0995.pres=81+566|u0995_u09CD.half_u0995.pres=84+566|u0995_u09CD.half_u0995.pres=87+566|u0995_u09CD.half_u0995.pres=90+566|u0995_u09CD.half_u0995.pres=93+566|u0995_u09CD.half_u0995.pres=96+566|u0995_u09CD.half_u0995.pres=99+566|space=102+213|u0995_u09B0_u09CD.blwf.vatu=103+643|u0995_u09CD.half_u09B2.pres=106+602]
fonts/sha1sum/1c2fb74c1b2aa173262734c1f616148f1648cfd6.ttf::U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0995,U+09CD,U+0995,U+0020,U+0995,U+09CD,U+09B0,U+0995,U+09CD,U+09B2:[u0995_u09CD.half_u0995.pres=0+566|u0995_u09CD.half_u0995.pres=3+566|u0995_u09CD.half_u0995.pres=6+566|u0995_u09CD.half_u0995.pres=9+566|u0995_u09CD.half_u0995.pres=12+566|u0995_u09CD.half_u0995.pres=15+566|u0995_u09CD.half_u0995.pres=18+566|u0995_u09CD.half_u0995.pres=21+566|u0995_u09CD.half_u0995.pres=24+566|u0995_u09CD.half_u0995.pres=27+566|u0995_u09CD.half_u0995.pres=30+566|u0995_u09CD.half_u0995.pres=33+566|u0995_u09CD.half_u0995.pres=36+566|u0995_u09CD.half_u0995.pres=39+566|u0995_u09CD.half_u0995.pres=42+566|u0995_u09CD.half_u0995.pres=45+566|u0995_u09CD.half_u0995.pres=48+566|u0995_u09CD.half_u0995.pres=51+566|u0995_u09CD.half_u0995.pres=54+566|u0995_u09CD.half_u0995.pres=57+566|u0995_u09CD.half_u0995.pres=60+566|u0995_u09CD.half_u0995.pres=63+566|u0995_u09CD.half_u0995.pres=66+566|u0995_u09CD.half_u0995.pres=69+566|u0995_u09CD.half_u0995.pres=72+566|u0995_u09CD.half_u0995.pres=75+566|u0995_u09CD.half_u0995.pres=78+566|u0995_u09CD.half_u0995.pres=81+566|u0995_u09CD.half_u0995.pres=84+566|u0995_u09CD.half_u0995.pres=87+566|u0995_u09CD.half_u0995.pres=90+566|u0995_u09CD.half_u0995.pres=93+566|u0995_u09CD.half_u0995.pres=96+566|u0995_u09CD.half_u0995.pres=99+566|u0995_u09CD.half_u0995.pres=102+566|space=105+213|u0995_u09B0_u09CD.blwf.vatu=106+643|u0995_u09CD.half_u09B2.pres=109+602]
fonts/sha1sum/a6c76d1bafde4a0b1026ebcc932d2e5c6fd02442.ttf::U+1004,U+103A,U+1039,U+101B,U+103D,U+102D:[uni101B103D=0+450|uni1004103A1039102D=0@-50,0+0]